Unicode and APIs

[Ken Raeburn]

On Sep 20, 2007, at 06:58, Sam Hartman wrote:
> I'd like to ask you to take a major step back here.  We seem to be
> disagreeing about some fundamental assumptions.
>
> I'm assuming that if krb5 knows the right characters then it will do
> the right thing regardless of what character set they are in.

Hm, okay.  I guess I was assuming (in the context of the previous  
discussion) that the context flag meant the library would deal with  
only one encoding at a time and the application would deal with  
conversions.  If you meant for the library to do the conversions (and  
know the local encoding), yeah, the API is still simpler in places.

krb5_rd_req could still be a problem if the local encoding is UTF-8  
but the client principal name on the wire is in some JS8 encoding.   
The config file might be a problem if we don't assume that it's  
always in the local encoding (which can vary from process to  
process), or always in UTF-8.

Ken